Google Gemma 4 12B
Run multimodal AI locally with an encoder-free architecture. An overview of Gemma 4 12B, a model designed to bring high-performance multimodal intelligence directly to your laptop. Gemma 4 12B processes text, vision, and audio natively without separate encoders, running on 16GB VRAM.
Reactions (1)
Similar products
by description similarity
Gemma 3
ai.google.devBuild with multimodal AI from Google
Gemma 3n
developers.googleblog.comRun powerful multimodal AI right on your phone
Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook
Google Gemma 4
blog.googleGoogle's most intelligent open models to date
Gemma Gem
github.comAI model embedded in a browser – no API keys, no cloud